Dec 15, 2009

rel=canonical part two

Google today officially announced support for cross-domain rel=canonicals.  This is a very important step.  I talked about rel=canonical when we announced it 10 months ago.  This tweak will be very useful in a variety of cases.  A few interesting ones:

  1. Any place where you want to move your site to a new domain but can't control server headers.  This can be really useful to help avoid lock in on free hosts for example.
  2. Companies that register several domain names as landing pages for offline advertising, a practice often used for tracking performance of the campaign, can now keep users on the original domain but keep spiders going to the right place so Google doesn't end up with 10 copies of the site.
  3. Syndicating content - if your syndication partner agrees - you can now make sure that your "link juice" flows back to the original article in Google.
I think we'll see lots of creative uses of this tweak in the future as we've seen with the original rel=canonical.  Let's just hope those creative talents are used for the power of good.


Jon Black said...

hey greg,

this is a great post. very helpful. my question is whether you feel that the rel=canonical link element can be used in place of an actual 301 redirect, or simply in conjunction with it?


Greg said...

I commented on this a little bit in my previous post on rel=canonical. There are cases where 301s make no sense (print friendly pages for example). There are cases where rel=canonical is also less than ideal - such as moving an entire website. don't you want users bookmarking the new site?

Obviously you can't actually use both techniques on the same page, but you both are useful in different situations.

Mr. Christopher said...
This comment has been removed by a blog administrator.
Greg said...

Christopher, I'd generally recommend against doing this multiple duplicate subdomains approach in the first place. I'm guessing that you are doing this to try to target search engines, not because it is a meaningful way for users to navigate your site. This is a frowned upon practice, see for reference the google webmaster guidelines: "Don't create multiple pages, subdomains, or domains with substantially duplicate content.".

That said, Google recommends that rel=canonical tags use absolute (full path) URLs. So, if you need these subdomains, and I'm understanding you correctly, I'd have have a rel=canonical tag to

Greg said...

I didn't realize that one of the previous comments had real URLs in them. I'm deleting the previous comment for this reason alone, and reposting it here with the URLs modified. If the author would like to contact me to discuss further, my email address, as published on this blog elsewhere, is I don't, as a rule, give out my @google address.

The comment was:

We have a situation on our site where we list different vendors for each state. The URL's are structured like and, etc. for all 50 states (and some cities also). But those subdomains are actually mirrored images of www. Meaning you can drop the alabama-weddings and stick in www and its the same thing.

The problem is that the user can dig down into that states vendors (/alabama/caterers.php or alabama/florists.php, etc.), so does the rel=canonical need to reflect the entire URL on every page or does it just need to reflect what the subdomain should be? Because if I just put in the rel=canonical on those pages, it's just going to redirect to the main page I'm assuming. So I may have answered my own question, but just checking because this seems like a root level thing and I'm just trying to see how the rel=canonical also applies to the multiple pages on a subdomain beyond just the main page.